Expression of a Human T-Cell Protein-Tyrosine-Phosphatase in Baby Hamster Kidney Cells (Phosphorylation/Regulation/Localization') DEBORAH E
Total Page：16
File Type：pdf, Size：1020Kb
Proc. Nati. Acad. Sci. USA Vol. 87, pp. 7280-7284, September 1990 Biochemistry Expression of a human T-cell protein-tyrosine-phosphatase in baby hamster kidney cells (phosphorylation/regulation/localization') DEBORAH E. COOL*t, NICHOLAS K. TONKSt, HARRY CHARBONNEAUt, EDMOND H. FISCHERt, AND EDWIN G. KREBS* *Howard Hughes Medical Institute and tDepartment of Biochemistry, University of Washington, Seattle, WA 98195 Contributed by Edwin G. Krebs, June 20, 1990 ABSTRACT A human T-cell cDNA encoding a 48-kDa their conserved predicted catalytic core, might imply that this protein-tyrosine-phosphatase (PTPase; protein-tyrosine-phos- region plays an important role in regulation. phate phosphohydrolase, EC 3.1.3.48) was cloned into a mam- The physiological functions of the low molecular mass and malian expression vector and introduced into baby hamster receptor-linked PTPases remain unclear, though obviously kidney cells, and stable colonies were isolated. The expressed they must be necessary to control the overall level ofprotein- PTPase was found to be associated with the particulate fraction tyrosine phosphorylation and thus would be required for of the cells, where it was essentially inactive in an in vitro assay regulating growth, differentiation, and transformation. In this unless first subjected to limited trypsinization; trypsin treat- present study, overexpression of the T-cell PTPase trans- ment generated an active fragment of 33 kDa by the removal fected into baby hamster kidney (BHK) cells was investi- of a carboxyl-terminal segment of the full-length enzyme. Gel gated to determine whether this would have adverse effects was associated on normal cell growth, including cytotoxicity. In addition, a filtration indicated that the expressed enzyme carboxyl-terminal truncated form of the enzyme closely with a complex of >600 kDa. Introduction of a premature stop resembling PTPase 1B was constructed. To clarify the pos- codon into the T-cell cDNA at position 1012 resulted in the sible function ofthe deleted segment, this was also expressed production of a fully active 37-kDa species that distributed in BHK cells. between both the particulate and soluble fractions. The trun- cated form of the enzyme was readily solubilized by detergents MATERIALS AND METHODS and was eluted within its predicted molecular mass range. Restriction and modifying enzymes were from Stratagene; These results suggest that the carboxyl-terminal segment is 125I-labeled protein A was from New England Nuclear; important in determining the localization and regulation of the platelet-derived growth factor (PDGF)j3 (c-sis) was from PTPase. The level of protein-tyrosine phosphorylation ob- Amgen Biologicals, and mouse monoclonal anti-phosphoty- served after 5 min ofplatelet-derived growth factor stimulation rosine antibody bound to agarose beads was from Oncogene was reduced in cells overexpressing either form of the phos- Science (Manhasset, NY). Conjugated goat anti-rabbit IgG phatase, indicating that both are active in vivo. Overexpressing horseradish peroxidase and the substrate color reagent were the truncated enzyme resulted in a growth rate that was purchased from Bio-Rad. Oligonucleotide primers and pep- approximately 50% of that observed in cells transfected with tides were synthesized by the Howard Hughes Biopolymer either the full-length PTPase cDNA or the vector alone. Synthesis Facility at the University of Washington. Plasmid Constructs and Site-Directed Mutagenesis. An A human T-cell cDNA encoding a 48-kDa protein-tyrosine- EcoRI-HindIII fragment [1.328 kilobase pairs (kbp)], com- phosphatase (PTPase; protein-tyrosine-phosphate phospho- prising the entire coding region of the PTPase cDNA and 60 hydrolase, EC 3.1.3.48) has been isolated (1). Its predicted and 22 bp of the 5' and 3' untranslated ends, respectively, to that of a were isolated from a human T-cell cDNA PTPase clone. amino acid sequence exhibited 75% identity Single-stranded ends of the cDNA fragment were treated 37-kDa human placenta enzyme (PTPase 1B) in a 236-residue with nuclease S1 and ligated to a 5.5-kbp Sma I fragment from conserved core segment found in all PTPases identified thus the pNUT expression vector (provided by R. Palmiter of the far (2-4). However, the T-cell PTPase also contained an University of Washington). The 5.5-kbp fragment encodes a 11-kDa extension at the carboxyl terminus, suggesting that dihydrofolate reductase cDNA under the regulation of a the placenta enzyme may have been isolated in a truncated, simian virus 40 (SV40) promoter and a Zn2+ metallothionein yet fully active, form. This was confirmed by the isolation of I promoter required for in vivo transcription of the newly cDNAs encoding either the human placenta enzyme (5, 6) or inserted cDNA (13). an isoform from rat brain (7), both with predicted molecular A mutation of the PTPase cDNA was performed as de- masses of approximately 50 kDa. scribed (14). An oligonucleotide (5'-GGGAACAGATAGAA- On the basis of the primary structure of placenta PTPase GAAG-3') identical to nucleotides 1004-1025 within the 1B (8), a family of PTPases has been established that includes T-cell PTPase cDNA except for a seven-base deletion was integral membrane proteins, such as CD45, with the struc- synthesized. It was used as a primer for in vitro DNA tural features of receptor molecules (9-12). The nature of the synthesis in a reaction with single-stranded phage M13 DNA ligands that might regulate the high molecular mass trans- containing the T-cell PTPase cDNA. A stop codon (TAG) membrane PTPases is unknown. Likewise, nothing is known was placed into the translation open reading frame following about the possible regulation of the low molecular mass class Arg-317. Selection of M13 phage that carry the altered cDNA of PTPases; however, the fact that their carboxyl extensions was performed by in situ plaque filter hybridization with display considerable sequence diversity, as compared with 32P-labeled oligonucleotide. The filters were washed under The publication costs of this article were defrayed in part by page charge Abbreviations: PTPase, protein-tyrosine-phosphatase; PDGF, plate- payment. This article must therefore be hereby marked "advertisement" let-derived growth factor. in accordance with 18 U.S.C. §1734 solely to indicate this fact. tTo whom reprint requests should be addressed. 7280 Downloaded by guest on September 29, 2021 Biochemistry: Cool et al. Proc. Natl. Acad. Sci. USA 87 (1990) 7281 conditions requiring perfect duplex formation for stability incubated on ice for approximately 20 min. The lysates were (15). A 1.6-kbp (Tha I and Ssp I) fragment encoding the centrifuged at 10,000 x g for 5 min at 4TC, and the protein mutated cDNA was isolated from the purified M13 recom- concentrations were determined as described by Bradford binant plasmid and ligated with the 5.5-kbp Sma I fragment (19). A suspension of agarose-linked monoclonal anti- from the pNUT expression vector described above. All phosphotyrosine antibody beads (30 1.l) was added to equiv- plasmid constructs were verified by DNA sequence analysis alent amounts of lysate protein in each immunoprecipitation, as described (16). and the mixture was rotated overnight at 4TC. The beads were Cell Culture and Transfections. BHK cells were routinely collected by centrifugation, washed twice in 20 mM Hepes, grown in Dulbecco's modified Eagle's medium containing pH 7.5/150 mM NaCI/0.1% Triton X-100/10% glycerol/200 10% (vol/vol) heat-inactivated fetal calf serum. The cells p.M orthovanadate, then twice again with the same buffer but were transfected with 10 ,ug of plasmid DNA by using the with increasing salt concentration to 0.5 M NaCl, and finally calcium phosphate precipitation method (17); after 24 hr, they with buffer containing 150 mM NaCl. The beads were boiled were switched to selection medium containing 250 AM meth- for 2 min in 30 p.l of Laemmli sample buffer (20). otrexate. Stable colonies were isolated at about 14 days The immunoprecipitated protein was subjected to Western posttransfection. For all experiments using PDGF stimula- blot analysis (21) with an anti-phosphotyrosine antibody tion, cells were first treated overnight with 80,uM ZnSO4, and kindly provided by J. Schlessinger, New York University then incubated for 48 hr in medium and 0.1% heat-inactivated Medical School. To detect antibody binding, 125I-labeled fetal calf serum for 48 hr. protein A (500,000 cpm/ml in 10 mM Tris, pH 7.4/150 mM Cell Homogenization. Cells grown to confluency were NaCl/1% bovine serum albumin) was added to the blot for 2 treated with 80 ,uM ZnSO4 for 12 hr, washed three times with hr and washed in 10 mM Tris, pH 7.4/150 mM NaCI/0.05% phosphate-buffered saline (PBS), scraped, and collected by Triton X-100. The blot was then subjected to autoradiography centrifuging at 800 x g for 5 min. The cell pellet was homog- for 2-5 days at room temperature. enized by using a Teflon Dounce homogenizer in low-salt buffer (LSB) (25 mM imidizole, pH 7.0/1 mM EDTA/1 mM RESULTS EGTA/0.1% 2-mercaptoethanol/2 mM MgCI2/0.002% phe- Expression of a Full-Length 48-kDa Human T-Cell PTPase nylmethysulfonyl fluoride/0.1 mM benzamidine/1 ,g of leu- in BHK Cells. The T-cell PTPase cDNA was transfected into peptin per ml/250 mM sucrose) and centrifuged at 5000 x g for BHK cells to determine the enzymatic properties of the 5 min. The supernatant was recentrifuged at 100,000 x g for 48-kDa form and to examine the physiological consequences 30 min at 4°C. All fractions, including the 5000 x g pellet (5P), ofits overexpression. Over 100 methotrexate-resistant stable the 100,000 x g pellet (100P), and the 100,000 x g supernatant colonies were observed of which 20 were selected and (100S) were assayed for PTPase activity.